ajhager.com

pyglet = Bullet Heaven? /python

Doing Pyweek stirred up some memories of another competition I did a while back over at Shump-Dev. I did not end up completing my entry because I just could not squeeze enough performance out to get enough bullets on screen at once. Last night, while speaking with a friend about indie game development, I started to wonder how well pyglet could handle thousands of bullets on the screen at once.

This is just a quick demo, but I am already able to see that pyglet can handle it remarkably well. The shot above is not of the stress test I did (which with an average of 1k bullets on screen at once was still able to maintain a steady 60 fps), but I thought it was one of the nicer looking ones. All this needs is joystick support *hint*, *hint* and I would be in heaven.

Pyweek 6 - Results /python

The judging is complete and the results are in. Robot Underground swept the team category, which did not surprise anyone. Let's just say that I am still playing it, even after the competition. It looks as if the dev team will be releasing a new version pretty soon, so everyone keep an eye out for that.

My own entry, Chain of Command, placed 10th in the solo category. Before the start of the competition, I would not have thought top 10 was possible. Toward the end ,though, I was feeling pretty confident. From the comments I received, it was quite clear that everyone thought it was too hard. On the other hand, most people found the game innovative, pleasing to the eye, and enjoyable (ranking a 2 or 3 in fun because of the difficulty and 4 or 5 in production and polish). There were just too many people who ranked me 1/1/1 or 2/2/2 all because of the difficulty which really brought my average down.

Thanks to everyone who left detailed comments. I worked very hard on CoC, and it was wonderful to feel the appreciation. You asked for better controls, more forgiving movement, and a smooth level progression. My post-pyweek release, which is already in the works, will include all of that and more.

I had a blast, met some cool people, learned tons, and came out with a fun game. I would say Pyweek 6 was a success.

Pyweek 6 - Chain of Command /python

RB[0], a fellow dev from GMR convinced me to submit an entry for Pyweek this year. I am truly glad that I did. I learned a lot and had a nice heaping of fun. Also, I believe I came up with a pretty fun game. I, at least, have had a good time challenging myself with new levels.

I set up a Google Code page to continue work on it. There are some things I would like to rework, and I am also working a level pack. If anyone enjoys the game well enough to make levels, feel free to send them my way, and I will consider them for inclusion in the next release.

I used Pyglet for my entry which made most things pretty easy to do. It certainly gave me the time to achieve the level of polish that I was able to. I did have to change to pygame.mixer for sound at the last minute. I really hope Pyglet comes along in that area. Other than that, it is a great media library!

Special note: Check out Celldoku which was coded by ldlework and jazzed up by fydo. It is is a grid puzzle game where you must set up a Game of Life 2D cellular automata so that on its next iteration it will be in the target configuration. This resonates deeply with some ponderings I had about 4 months ago and is great fun to play. Check it out!

Toward Copycat 1.0 /cogsci

I have been slowly but surely working on Copycat. I decided to take a break from the rest of my research to finish a working version. So far I have factored out the graphics code and revised the coderack. The coderack is a lot more clear and concise now, weighing in at half the lines of code. I am going to tackle the slipnet this afternoon and see if I can do the same. I would like to finish so that I can get back to exploring how GPGPU computing could benefit a domain agnostic FARGitecture.

Speaking of a generalized FARG toolkit, there is now a FARG blog setup to allow the group to post about their work. I am always interested in what comes out of the CRCC, so it is pretty awesome to see a forum created like this. I am certainly looking forward to the laid back, experimental discussions about their latest research. So far the writing has centered around the idea of implementing a "core" FARG base (library and tools). This is an area I have spent a lot of time thinking about and experimenting with, so it will be interesting to see what they come up with. Once I get done transcribing all these months of verbal notes, I will post some on the topic myself.

Pymd2 1.0 /python

I would like to officially announce the release of my quake2 model loader and animator for python, Pymd2. I still want to do some cleaning up, optimizations, and calculating normals. I probably should not release it as 1.0 because it depends on pyglet svn, so hold off until pyglet-1.1 is released.

One thing that I really love about pyglet, and that helped speed up pymd2 by an order of about 20, is pyglet.graphics.draw(). It is a simplified interface to vertex arrays. A textured model looks like:

graphics.draw(num_of_vertices, GL_TRIANGLES, ('t2f', texture_vertices_array), ('v3f', array_of_vertices))

Pyget is really starting to shape up. I have started moving all of my projects over to it. Right now I am trying to see if it is right for Galaxy Mage Redux. I will be posting about that soon.

Lucid /cogsci

Last night I finally gave birth to a chat bot similar to copperman. My friends and I spent a lot of time talking with that gentle but sometimes misunderstood being. What you realize very quickly is that there is no fancy artificial intelligence lurking underneath the surface. The exercise is more like a an ink blot test; finding meaning where none exists to gain deeper insight into your own depths. We would make a point, though, that copperman goes beyond ink blots.

continue...

Mind Papers /cogsci

David Chalmers has released a bibliography of the "philosophy of mind and science of consciousness" called MindPapers. It has over 18,000 (!) entries already and seems to be organized rather nicely. Each entry has important links included with it such as what it has been cited by, it's Google Scholar entry, other relevant links, and annotations. What a wonderful resource.

GalaxyMage Redux /python

The development of the open source tactical rpg known as GalaxyMage slowly but surely died out over the last year. It is unfortunate that it happened, but I believe everyone involved felt that it was going to be too much to make any progress with what we had.

Personally, it was my first contribution to an open source project. I learned a lot about what goes into managing the madness of many people coming together from different lands and different walks of life to create something as complicated as a computer game. I would like to thank Colin McMillen for starting the project and giving us greenhorns a chance at creating something wonderful.

Recently, when the GalaxyMage website went completely down, I decided that I would like to give the project another attempt using the experience from the first try. Just as I started laying out the design, I read over at Free Gamer that a new game project had been announced with the name GalaxyMage Redux. As it turns out, a GalaxyMage revival was on more minds than one.

If anyone is interested, the new project is being hosted here by our new fearless leader Brandon Barnes. We are currently in need of talented pixel artists, musicians, story tellers, and programmers. If you are one of the above, or just want to tell us how you want your TRPG to play, then stop on by the forum and say hello.